Author: Eduard Nemeth
Date: 23:31:57 03/12/03
Go up one level in this thread
On March 12, 2003 at 22:04:08, Uri Blass wrote: >On March 12, 2003 at 20:38:46, Mark Young wrote: > >>On March 12, 2003 at 15:25:41, Uri Blass wrote: >> >>>On March 12, 2003 at 14:37:17, Mike Hood wrote: >>> >>>>On March 12, 2003 at 11:43:29, Thorsten Czub wrote: >>>> >>>>>On March 12, 2003 at 11:35:24, Eduard Nemeth wrote: >>>>>>Ich möchte Frans Morsch schreiben (wegen betatets). >>>>> >>>>> >>>>>Fritz is not tested in beta-tests. the customer is the beta-tester :-) >>>> >>>>As far as I know, new GUIs are put through a beta test process, but the engines >>>>are not beta-tested, on the assumption that the programmer of an engine knows >>>>best how to test it himself. >>> >>>I do not understand >>> >>>1)It seems to me that the assumption that the programmer of an engine knows best >>>how to test the engine does not contradict the conjecture that beta testers may >>>help the programmer to develop a better version. >>> >>>2)It seems to me that the programmer has not infinite computer time and beta >>>testers may use computer time to help him. >>> >>>3)It seems to me that there are also cases when beta testers give productive >>>ideas for chess programs. >>> >>>Even if the programmer knows better it does not mean that other people cannot >>>contribute. >>> >>>Can you explain what is the error in my assumptions or in my logic and why do >>>you think that the best that I can do is never to send movei to beta testers? >> >>As far as I know Fritz has been the strongest engine over time then any other >>chess program, looking at the SSDF list. >> >>I think Fritz has held the Top Spot more and longer then any other program. If >>this is true why would he what to change? It seems to me he does know best, and >>his testing methods have been better then other programmers. Fritz's results >>prove it. > >1)Fritz results prove that fritz is always best or one of the best. >It does not prove that the testing methods are the best. > >2)I did not talk about the testing methods of Fritz. >I did not claim that the engine is not tested by beta testers. > >Even if it is the case I have no way to know it. >I only know that other programmers use beta testers. > >Uri before Fritz 7 came to the sales, I asked one on the Fritz server the programmer like it for Fritz 7 stand at that time. I said literally: are you reliably that Fritz 7 (engines) well succeeded? It answered me: we are completely safe us. But then we found some BUG in the engine Fritz 7,001 (I) to Fritz 7,002 (some together) and in Fritz 7,006 (I). And now I discovered again a BUG in Fritz 8. Germal original: bevor Fritz 7 zum Verkauf kam, fragte ich damals auf dem Fritz-Server einen der Programmierer wie es um Fritz 7 steht. Ich sagte wörtlich: "seid ihr sicher dass Fritz 7 (Engine) gut gelungen ist?" Er antwortete mir: "ja wir sind uns ganz sicher." Doch dann fanden wir einige grobe BUGs in der Engine Fritz 7.001 (ich) Fritz 7.002 (einige zusammen) und in Fritz 7.006 (ich). Und nun habe ich wieder einen BUG in Fritz 8 entdeckt. Eduard Eduard
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.